Models and Usage

Choose a Bkper AI model, compare usage rates, and track the monthly allowance included with your plan.

Bkper AI models power Bkper Agent, Bkper CLI Agent, and the apps you use in Bkper. They come with one model catalog and one monthly allowance, included with your plan. You need no model-provider account or API key.

Choose a model

In Bkper Agent, pick the model below the message box, or type /model. New sessions start with GPT Sol on paid plans and GPT Luna on the Free plan.
In Bkper CLI Agent, type /model.

A high-capability model, such as GPT Sol, suits complex reasoning and long reports. A fast, economical model, such as GPT Luna or Gemini Flash-Lite, suits quick questions and many small tasks. The usage rates show what each costs against your allowance.

Decision models, such as Jev, are not for conversations. Apps ask them for typed judgments, such as which Account fits a transaction; see Decision Models.

This page does not apply when you connect Bkper to ChatGPT or Claude and keep using the models provided by those products.

Models

Public IDs identify stable Bkper model families. Bkper can upgrade the concrete provider revision behind a family without changing its public ID, saved client configuration, or usage-report identity.

Model	What it offers
Claude Sonnet 5.5 `claude-sonnet` Language	Fast, capable multimodal model for coding, reasoning, and agent workflows. Reasoning: `low`, `medium`, `high` Limits: 300k context · 64k output
DeepSeek V4.1 Flash 1017 `deepseek-flash` Language	Fast, economical multimodal reasoning, coding, and long-context work. Reasoning: `high`, `max` Limits: 300k context · 64k output
Gemini 3.5 Flash-Lite `gemini-flash-lite` Language	Economical multimodal model for fast, high-volume tasks and tool use. Reasoning: `minimal`, `low`, `medium`, `high` Limits: 1,048,576 context · 65,536 output
Gemini 3.8 Flash `gemini-flash` Language	Fast multimodal reasoning and tool use with balanced usage rates. Reasoning: `low`, `medium` Limits: 1,048,576 context · 65,536 output
GPT-6 Luna `gpt-luna` Language	Cost-efficient GPT model for fast, high-volume workloads. Reasoning: `high`, `xhigh`, `max` Limits: 272k context · 64k output
GPT-6.1 Sol `gpt-sol` Language	High-capability GPT model for complex reasoning, coding, and demanding agent workflows. Reasoning: `low`, `medium`, `high` Limits: 272k context · 64k output
Grok 4.7 `grok` Language	General-purpose model for chat, coding, and agentic tool use. Reasoning: `low`, `medium` Limits: 200k context · 32k output
GLM 5.3 Flash `glm-flash` Language	Fast, economical multimodal reasoning, coding, and long-context work. Reasoning: `low`, `high`, `max` Limits: 1,048,576 context · 64k output
Jev 1.13 `jev` Decision	Fast typed judgments for noul, choice, and score questions. Question types: `noul`, `choice`, `score` Limits: 65,536 context · 32,768 state + question

Bkper CLI loads the current context window, output limit, and reasoning profile for language models from the live catalog. Compatible clients can request lower output budgets and any listed reasoning effort. Anthropic safeguards can refuse Claude Sonnet requests; a tool call is not guaranteed.

Jev evaluates one shared state of text, JSON, and images against independent questions in parallel. Its responses report the concrete Jev revision in model; the catalog and usage history keep the stable jev ID. They are complete JSON and do not stream or create conversation state. See Decision Models for requests, responses, and decision patterns.

How usage works

Bkper AI is included with eligible plans and controlled through one monthly allowance:

Bkper Agent and Bkper CLI Agent connect with your Bkper login; no separate provider setup is required.
Model usage reduces the included allowance and is not billed separately.
New requests stop when the recorded allowance is exhausted; there are no automatic paid overages.
External providers remain available where supported and do not consume the Bkper AI allowance.

Requests already in flight may settle after the allowance check, so recorded usage can slightly exceed the limit under concurrency. Bkper does not bill that difference as an AI overage.

Usage rates

Usage rates reduce the included monthly allowance. They are not billed separately by Bkper.

USD of included usage per one million tokens

Model	Input	Cache read	Cache write	Output
Claude Sonnet 5.5	$3.00	$0.15	$3.80	$15.00
DeepSeek V4.1 Flash 1017	$0.45	$0.009	$0.00	$1.80
Gemini 3.5 Flash-Lite	$0.45	$0.045	$0.00	$3.80
Gemini 3.8 Flash	$2.30	$0.23	$0.00	$11.00
GPT-6 Luna	$0.15	$0.015	$0.19	$0.75
GPT-6.1 Sol	$3.00	$0.15	$3.80	$15.00
Grok 4.7	$3.00	$0.75	$0.00	$9.00
GLM 5.3 Flash	$0.23	$0.045	$0.00	$0.75
Jev 1.13 (decision)	$0.063	—	—	$0.00

Input means tokens sent without a cache match. Cache read means reused input already stored by the provider. Cache write means input added to a provider cache. Output includes generated response and reasoning tokens reported by the provider. Jev is input-only, so cache rates do not apply and returned typed answers add no output-token rate.

Claude Sonnet refusals are reported separately from successful responses. Documented unbilled pre-output Anthropic refusals retain reported token counts but consume no allowance; other refusals may consume allowance.

Monthly allowance

For paid plans, the monthly AI allowance equals the normalized monthly software subscription value:

Monthly plans use the recurring monthly software subscription value.
Annual plans divide the recurring annual software subscription value by 12.

Only recurring software subscription value counts. Professional services, implementation, consulting, taxes, one-time charges, credits, refunds, and prorations do not increase the allowance. Free users receive a separately configured trial allowance.

The allowance resets monthly and unused value does not roll over. It is an inference entitlement—not cash, refund value, or transferable credit. The authenticated Bkper AI usage dashboard shows your exact current allowance.

Individual and pooled usage

Allowance scope follows the subscription:

Free and Standard usage is assigned to the individual user.
Business usage is pooled across the subscribed domain.
Everyone sharing a pooled allowance reduces the same monthly total.

A pooled allowance does not make every user’s request history visible to everyone. Visibility depends on the viewer’s billing role.

Usage visibility

The Bkper AI usage dashboard separates allowance visibility from request attribution:

Regular users see the shared allowance remaining and their own requests and usage.
The billing or subscription administrator sees domain-wide usage attributed by user, AI model, and app or source.
The dashboard does not expose prompts or responses.

Usage value is an estimate based on the published rates above. It shows how much of the included allowance a request consumed; it is not a separate Bkper charge.

When the limit is reached

Bkper AI blocks new allowance-backed requests once the recorded monthly allowance is exhausted. There are no automatic paid Bkper AI overages at launch.

Bkper AI is not a lock-in: where supported, you can connect an external model provider at any time. External subscriptions, API keys, charges, privacy terms, and limits are governed by that provider and do not use the included Bkper AI allowance.

Privacy, retention, and caching

Zero data retention where available. Minimum retention everywhere. Bkper disables provider application storage on every request and keeps prompt and response content out of usage logs.

Bkper applies these boundaries across the inference path:

omitted store becomes false, and store: true is rejected before provider dispatch;
Bkper usage records contain attribution, status, token, cache, and cost metadata—not prompt or response content;
detailed Bkper usage events are retained for 180 days, while daily aggregate usage remains available for allowance enforcement and reporting;
Cloudflare AI Gateway payload logging is disabled, while content-free request metadata remains available for observability;
exact, non-streaming structured JSON requests and responses sent through POST /v1/responses may be cached by Cloudflare for up to 24 hours;
decision results sent through POST /v1/decisions are cached by Bkper for 15 minutes, keyed by a hash of the user and the exact request; the cached result contains the answers, including choice option names and score level text, but not the state or instructions;
provider-native prompt caches may hold language-model content temporarily under the selected provider’s caching policy.

Provider retention controls differ:

Provider	Bkper configuration	Provider retention boundary
xAI	`store: false`; Zero Data Retention is active for Bkper’s xAI team	Prompt and response retention is disabled under the enabled Zero Data Retention control.
OpenAI	`store: false`; Bkper organization API-call logging is disabled	API data is not used for training by default. Customer content may remain in abuse-monitoring logs for up to 30 days because Bkper does not currently have OpenAI Zero Data Retention.
Google Gemini paid API	`store: false`; Bkper avoids Search and Maps grounding, the File API, and explicit context caching	Paid API content is not used for training. Content may be retained for limited abuse monitoring because Bkper’s AI Studio project does not currently have approved Zero Data Retention. Gemini may also use project-isolated in-memory caching for up to 24 hours.
Fireworks AI	`store: false`; Zero Data Retention is active by default; used as the current serving route for open-weight models	Fireworks does not log or persist prompt or generation data for open models without explicit opt-in. Prompt caching may retain data in volatile memory for several minutes. `store: false` prevents Response API conversation storage.
TypeSafe	Decision requests use Bkper’s server-side TypeSafe credential; Bkper usage logs exclude state and answers	TypeSafe states that customer requests and responses are not used for training. Retention and Zero Data Retention availability follow TypeSafe’s current service terms.

Caching does not create a retrievable conversation or add prompt content to usage logs. If you set prompt_cache_key, use an opaque identifier: it may appear in the usage dashboard.

Provider policies and Bkper configurations can change. Review these primary references for the current boundaries:

How Bkper selects language models

We build Bkper with the Bkper CLI Agent and use it every day. We test many language models through real work and include only those that consistently work well for us within our cost and control constraints. The catalog is a practical, opinionated shortlist—not a directory of every available model.

Bkper prioritizes strong results at controlled cost — the efficient frontier of capability per dollar — rather than pursuing the highest benchmark score at any price. Selection also considers:

results and reliability in daily agent workflows;
effective model capabilities and tool use;
observed usage cost;
model capabilities and controls;
public benchmarks.

Explore the live DeepSWE leaderboard.

DeepSWE measures long-horizon software-engineering work. It is one input into model selection, not a measure of accounting accuracy or a guarantee of performance in Bkper workflows.

Model creators

Model	Type	Creator
Claude Sonnet 5.5	Language	Anthropic
DeepSeek V4.1 Flash 1017	Language	DeepSeek
Gemini 3.5 Flash-Lite	Language	Google
Gemini 3.8 Flash	Language	Google
GPT-6 Luna	Language	OpenAI
GPT-6.1 Sol	Language	OpenAI
Grok 4.7	Language	xAI
GLM 5.3 Flash	Language	Z.ai
Jev 1.13	Decision	TypeSafe

Sources

Last synchronized: 2026-10-23

Models.dev provider assets are provided under the MIT License. Model creator names and logos remain trademarks of their respective owners.

Next steps

Use Bkper Agent in your Books.
Use Bkper CLI Agent on your own computer.
Ask decision models for typed judgments.
Open the Bkper AI usage dashboard.
Building an app? Add Bkper AI to an App, or call the models from your own code with the AI Gateway API.
Back to AI overview.